Head-to-Head Model Showdown
Mistral Large 3 vs Qwen3.5-397B-A17B
Mistral Large 3 vs Qwen3.5-397B-A17B side by side: arena.ai text score, list price per 1M tokens, context window and licence, with sources. Data checked 2026-10-11.
Executive Verdict
On the arena.ai text leaderboard, Mistral Large 3 scores 1414 (±3, #143) and Qwen3.5-397B-A17B scores 1442 (±3, #92). Qwen3.5-397B-A17B leads by 28 points, outside the margin of error. Mistral Large 3: $0.50 input / $1.50 output per 1M tokens. Qwen3.5-397B-A17B: $0.50 / $3.60. Blended at 3 input : 1 output, Mistral Large 3 costs 41% less ($0.75 vs $1.27 per 1M tokens). Both accept 256K tokens. Both publish open weights (Apache 2.0; Apache 2.0), so both can be self-hosted.